“FRee trade is almost dead,” said Morris Chang, founder of tsmc, dampening spirits at an event in December to celebrate a milestone in the construction of the Taiwanese chipmaker’s new Arizona factory. The remark was not uncharacteristic. In July, he challenged America’s efforts to adopt chip manufacturing hometown an “exercise in futility”. Until recently, rich-world governments mostly shared his sentiment. But concerns about supply chain security in a strained world encourage experimentation. History offers some reasons for optimism – but also many for concern.
Industrial policy is about as old as industry itself. No sooner had the Industrial Revolution begun in Britain than Alexander Hamilton, America’s first Treasury Secretary, advocated protecting his country’s industry, declaring that Adam Smith’s arguments for free trade were “although ‘geometrically true’ ‘practically false’”. America, France and Germany industrialized behind customs barriers. After World War II, numerous governments tried to kick-start industrialization, with apparent success in countries like Japan and South Korea and rather mixed results elsewhere. Politics today is of a different nature: pursued by countries already at the technological frontier, in a world of complex global supply chains. But previous research still offers valuable lessons.
Recent interventions are mainly based on “infant industry” arguments. The idea is that if government corrects a market failure, a particular industry could thrive on its own in an economy where it is emerging or absent. Local businesses may need investments in know-how or equipment to be competitive that imperfect capital markets cannot fund. Alternatively, production could require a network of suppliers and manufacturers, but companies struggle to coordinate. Or there may be information problems. An economy may have undiscovered potential, but an entrepreneur who seeks it risks revealing it to competitors, costing him the opportunity to capitalize on his discovery. In any case, government support or brief protection from foreign competition (or both) could create the space the industry needs to mature.
Figuring out whether these theories are practically or only geometrically true is no easy task. Industrial policy is never pursued in isolation, which means it is often difficult to isolate its impact. Still, careful work suggests that infant industry policies can work in the real world. For example, in the 1970s America was the dominant exporter of computer chips. The Japanese government invested heavily in semiconductor research and may have helped Japanese chip-consuming companies coordinate to source most of their supply from young Japanese manufacturers (thus shutting out American companies from the market). The work of Richard Baldwin of the Graduate Institute in Geneva and Paul Krugman of the City University of New York concluded that this policy aided the accumulation of know-how, without which Japanese firms would never have been successful in export markets.
Recent work by Harvard University’s Myrto Kalouptsidi showed that Chinese shipyard subsidies reduced costs by up to 20% between 2006 and 2012. She believes these subsidies contributed to a major reallocation of shipbuilding, with Japan being the big loser. Other research highlights other cases where interventions have helped industry to establish itself in the market and have had a significant impact on the global distribution of production. At least sometimes, a comparative advantage can be constructed.
Still, a lot of caution is in order. Interventions often increase costs and thus harm consumers. Messrs. Baldwin and Krugman felt that efforts to build a chip-exporting industry made the Japanese worse off overall. Because the output of one industry is often the input for another, helping upstream producers can disrupt the supply chain. Bruce Blonigen of the University of Oregon examined efforts to boost the steel industry in 21 countries and found that such intervention severely hampered the export competitiveness of downstream industries.
Governments, for their part, must be willing to stop aid so that eventually the winners swim while the losers sink. Otherwise, zombie companies tie up capital and labor and slow down growth. Local conditions play a role. A study by EU Investment fund for poorer regions, by Sascha Becker from the University of Warwick and by Peter Egger and Maximilian von Ehrlich eth Zurich, found that cash led to faster growth in investment and income – but only in places with strong institutions and well-educated workers.
And how the world is rediscover, lax policies can provoke retaliation, making everyone worse off. This can be a particular problem at a time when demanding goods are being produced along cross-border supply chains. If friendly countries don’t coordinate, they could end up funding duplicate factories that can’t all be economical, or orphan industries without access to the foreign components they need to be competitive.
hard problems
Measures that fill institutional gaps are safer. Dartmouth College’s Douglas Irwin notes that America’s tariffs in the 19th century do not appear to have been instrumental in its rise to industrial dominance. More important were banking laws that made it easier to save and invest. In their survey, Ann Harrison and Andrés Rodríguez-Clare from the University of California, Berkeley doubt that “hard” interventions that distort market prices have any benefit, but see an important role for “soft” cooperation between companies and the state solve coordination errors.
That doesn’t mean the “tougher” parts of America’s policy mix will derail his reshoring venture. For his part, Mr Chang insisted in December that he made his comments “with the full expectation that we shall succeed.” Indeed, the most pressing concern may be less that America’s gambit will fail than that it will succeed in boosting domestic industry – and in doing so make a fragmented world look worse. ■
